the rapid expansion of data worldwide invites the need for more distributed solutions in order to apply machine learning on a much wider scale. the resultant distributed learning systems can have various degrees of ce...
详细信息
ISBN:
(数字)9781665499583
ISBN:
(纸本)9781665499583
the rapid expansion of data worldwide invites the need for more distributed solutions in order to apply machine learning on a much wider scale. the resultant distributed learning systems can have various degrees of centralization. In this work, we demonstrate our solution FLoBC for building a generic decentralized federated learning system using the blockchain technology, accommodating any machine learning model that is compatible with gradient descent optimization. We present our system design comprising the two decentralized actors: trainer and validator, alongside our methodology for ensuring reliable and efficient operation of said system. Finally, we utilize FLoBC as an experimental sandbox to compare and contrast the effects of trainer-to-validator ratio, reward-penalty policy, and model synchronization schemes on the overall system performance, ultimately showing by example that a decentralized federated learning system is indeed a feasible alternative to more centralized architectures. [GRAPHICS]
the decentralized, distributed ledger, a fundamental aspect of blockchain technology, plays a pivotal role in various sectors, including electronic voting (e-voting) systems. When considering e-voting, the significanc...
详细信息
Aiming at the problem that the different driving modes of the four motors of distributed drive electric vehicles affect the vehicle energy consumption, the driver driving intention recognition mechanism based on fuzzy...
详细信息
With biometric identification systems becoming increasingly ubiquitous, their complexity is escalating due to the integration of diverse sensors and modalities, aimed at minimizing error rates. the current paradigm fo...
详细信息
Static itinerary planning is a commonly employed multi-hop itinerary planning method in wireless sensor networks (WSN) due to its commendable energy efficiency and ability to minimize overhead. However, this approach ...
详细信息
Binary neural network (BNN) is widely used in speech recognition, image processing and other fields to save memory and speed up computing. However, the accuracy of the existing binarization scheme in the realistic dat...
详细信息
Owing to the difference of capacity and initial state of charge (SOC) for the multi-group energy storage units, as well as the different line resistance, the SOC of the energy storage unit cannot be equalized under th...
详细信息
Reputation systems have been one method of solving the unique challenges that face distributed networks of independent operators. Fundamentally, historical performance must be considered in a way that attempts to pred...
详细信息
ISBN:
(纸本)9781665474085
Reputation systems have been one method of solving the unique challenges that face distributed networks of independent operators. Fundamentally, historical performance must be considered in a way that attempts to predict future behavior, optimize present functionality, and provide some measure of immutable recording. In this paper, a three-part system, MnemoSys, is proposed to solve this diverse set of problems. First, historical performance is dynamically weighted and scored using geometrically expanding time windows. Second, a quorum is abstracted as a restricted Boltzmann machine to produce a conditional probability estimate of log-normal likelihood of good-faith behavior. third, all rewards and punishments are recorded on an immutable, decentralized ledger. Our experimentation shows that when applied iteratively to an entire network, consistently under-performing nodes are removed, network stability is maintained even with high percentages of simulated error, and global network parameters are optimized in the long-term.
As people have higher requirements for object detection accuracy and adaptability, this research proposes a single-stage algorithm for thermal image target detection based on Yolo v4. the algorithm combines the visibl...
详细信息
Military tactical scenarios have been shifting to more often consider combat situations in urban environments. threats in these environments are generally more dynamic in nature, imposing new requirements on sensors a...
详细信息
ISBN:
(纸本)9798350342734
Military tactical scenarios have been shifting to more often consider combat situations in urban environments. threats in these environments are generally more dynamic in nature, imposing new requirements on sensors and communications systemsthat support military operations. Wireless sensor networks (WSNs) with a large number of small and mobile computing nodes became the typical solution. However, WSNs demand additional complexity to dynamically manage their tasks, resource allocation, mobility, power consumption, and communication. this paper illustrates the integration of AI techniques into a Battle Management System (BMS) to support military operations in urban environments. the BMS is enhanced with an AI-based planner able to plan tasks, allocate resources, and monitor the WSN operation. the planner takes into consideration energy harvesting capabilities, secure data transfer, and authorization procedures. It generates plans using the information received from the sensors. In case new situations emerge, based on data fusion information, it automatically replans to adapt to the uncertainty in the environment. Finally, it takes into account the coverage between the different components to optimize the communications and better support WSN's operator(s) and their activities.
暂无评论